Forecasting is always somewhat of an estimation for the future, even when based on advanced mathematics and incorporating other data and sources to improve accuracy.

This post is not intended to delve deeply into the maths, but rather to get your thoughts or feedback on some of the influences you have seen, especially the outliers. We have all seen the impact COVID has had on modelling, with a combination of voluntary reduction in travel from passengers as well as mandatory restrictions from governments. However, over the years we have also seen some interesting items impacting on demand. Some of these appear to make sense, where others seem to fly in the face of logic.

Day to day weather forecasts, can have regular impact on traffic, especially on day outing type passengers. A spur of the moment trip may be taken if the weather forecast for the coming weekend is extremely good, or conversely a really bad forecast may cause late cancellations and/or no shows.

Strike action, natural disaster (or catastrophic weather), can sometimes lead to a negative demand impact, whereas something as simple as a series of concerts or sporting events often results in increased demand to the destinations of those venues.

Sometimes, however, the impact can be counter intuitive. Many years ago, working with one of our clients, one of their destinations had volcanic eruptions nearby. This was not on the scale of Eyjafjallajökull’s eruption in 2010, which caused chaos in Europe for weeks, but was still a significant event. My thinking, on a personal level, was “This will lead to a substantial decrease in demand on those routes for a time”. However, in looking at the data, it turns out that there was a marked increase in traffic to the nearby destinations as well as on feeder routes. As it turns out, scientists from around the globe were rushing there to do scientific research, intentionally putting themselves in harm’s way to advance our understanding of nature. This has now become a somewhat anecdotal part of our forecasting training, asking people to look at the data rather than an expectation of human behavior.

Going back to Eyjafjallajökull’s eruption this had different impacts on different clients/Industries. The airlines clients were down who simply could not fly, but there were positive impacts elsewhere, for example in different industries are impacted such as C/F and Channel Tunnel where demand picked up due to no fly.

What is the strangest thing you have seen which impacts demand in completely unexpected fashion?